Bootstrap-GS: Self-Supervised Augmentation for High-Fidelity Gaussian Splatting

TL;DR

This paper introduces Bootstrap-GS, a self-supervised augmentation method for 3D Gaussian Splatting that enhances rendering quality and addresses sampling deficiencies, leading to fewer artifacts and better metrics.

Contribution

It presents a novel bootstrapping framework that synthesizes pseudo-ground truth views to improve 3D-GS training and performance.

Findings

Reduces artifacts in novel view synthesis

Improves quantitative rendering metrics

Enhances adaptability of Gaussian-based methods

Abstract

Recent advancements in 3D Gaussian Splatting (3D-GS) have established new benchmarks for rendering quality and efficiency in 3D reconstruction. However, 3D-GS faces critical limitations when generating novel views that significantly deviate from those encountered during training. Moreover, issues such as dilation and aliasing arise during zoom operations. These challenges stem from a fundamental issue: training sampling deficiency. In this paper, we introduce a bootstrapping framework to address this problem. Our approach synthesizes pseudo-ground truth from novel views that align with the limited training set and reintegrates these synthesized views into the training pipeline. Experimental results demonstrate that our bootstrapping technique not only reduces artifacts but also improves quantitative metrics. Furthermore, our technique is highly adaptable, allowing various Gaussian-based method to benefit from its integration.